Reactions to positive deviance: Social identity and attribution dimensions

This research examines whether evaluations of positive deviates (i.e. high achieving group members) are influenced by the attributions they make for their performance. We argue that ingroup positive deviates who make group attributions help enhance the ingroup's image and thus attract favorable evaluations. In Experiment 1, ingroup positive deviates who made group attributions were generally evaluated more favorably than ingroup positive deviates who made individual attributions. There was also evidence that the positive deviates' attribution style influenced group and self-evaluations. Evaluations of outgroup positive deviates were not influenced by their attribution style. In Experiment 2, an ingroup positive deviate who was successful and attributed that success to the group was upgraded relative to an ingroup positive deviate who made individual attributions. Group evaluations were also higher when the positive deviate made group attributions. This pattern did not emerge when the positive deviate failed. The results are discussed from a social identity perspective.

Authorship Attribution with Support Vector Machines

In this paper we explore the use of text-mining methods for the identification of the author of a text. We apply the support vector machine (SVM) to this problem, as it is able to cope with half a million of inputs it requires no feature selection and can process the frequency vector of all words of a text. We performed a number of experiments with texts from a German newspaper. With nearly perfect reliability the SVM was able to reject other authors and detected the target author in 60–80% of the cases. In a second experiment, we ignored nouns, verbs and adjectives and replaced them by grammatical tags and bigrams. This resulted in slightly reduced performance. Author detection with SVMs on full word forms was remarkably robust even if the author wrote about different topics.